School cost and quality....again

There's really nothing new in this Herald article on a consultant's analysis of school costs in Vermont, which was presented to a joint meeting of the House and Senate Education Committees.  The consultant's analysis was done earlier this year and has been on the legislature's website (here, for the short version, and here for the detail) since July.   And Hugh Kemper commented on the same topic on these pages earlier this year.  Still, there's always something to think about when it comes to education spending in Vermont.

Reading the article carefully makes me think back to the problems the Soviet Union had as production managers in Moscow tried to figure out why their production targets for cost and quality were not being met:

Legislators had asked economists Thomas Kavet and Nicholas Rockler to "drill down" into the causes behind education spending and figure out why some schools are more expensive and what taxpayers are getting for their money.

It can't be done without a large-scale effort to reconcile and sort data, Kavet said Thursday. That is because below the supervisory union level there is little consistency in how different expenses are allocated and accounted for. So in one supervisory union, special education costs are counted as a supervisory expense, in another they are considered a school-by-school expense.

Rather than focusing on the product itself, the focus is on getting more and more detail about what's going on, not just in each district, but in each school.

We as consumers  don't need to know all those details, nor should we want to.  We look for a good or service and we decide on its price/quality attributes and either buy it, or buy from some other producer.   To make a crude, but appropriate analogy, we look at cars and decide whether we want to buy a Ford Taurus or Toyota Camry.  It's up to the management of the two firms to figure out how best to produce the car.  If we had a car czar, that person would have the unenviable job of figuring out how to produce the "best" car for Vermonters. 

The analogy is that it is the principals and superintendents who should be figuring this stuff out, and what should hold their feet to the fire is not some mandate from Montpelier, but the fear that if they are doing a bad job they'll lose students and money, and if they're doing a good job, they'll gain students and money.  And these education managers can then reward their teachers (and themselves) for doing a good job or get rid of the ones who are not.

What's missing in this whole discussion of getting at the root causes of Vermont's high education costs is the total absence from the discussion of the role of the ultimate consumer of education services.  Where are the parents in this analysis?   They have almost no choice in what is being provided to their children in their local schools.   Is it any wonder that quality is unclear, and costs are out of control (insulating taxpayers from costs is another problem that we've talked about here).

I know, people will say that they do have input, through their local school board.  As most people who have served on their local boards know (and I have), that is not true.   And the only way parents can choose to buy some other form of education for their kids is to either move to another district (a costly undertaking) or send their kids to private school (even costlier).

It appears that the consultants would like to see an education czar (having an education commissioner is not czar-like enough, I guess):

Making it [evaluating costs] more difficult is the fact that there is no office or person in state government who oversees how money is spent — even though much of the money that supports schools is now gathered through a statewide property tax system.

"There is not any cohesive oversight of the big picture," Kavet [the consultant] said. "The financing is done that way, but the operation is the sum of all of these little parts."

So what is needed, then is more centralization of information.  And with more information, the people in charge will make a better decision.  It didn't work in the Soviet Union, and it won't control the costs of education. 

Nor will it improve quality.  We always pat ourselves on the back when it comes to evaluating educational quality in the state, and the consultant parroted the usual phrases:

That is important, if the state is to be able to evaluate how well it is spending its education money, he said. Although Vermont students do well in national standardized testing, the state also spends more per pupil than the vast majority of states and has the lowest pupil to teacher ratio in the nation.

The quality is not there though, when you compare Vermont's test scores to the non-minority test scores nationally.  We are just about average.  Since Vermont has about 2% minority students and the U.S. minority population is about 30%,  the relevant comparison is between  Vermont students and the U.S. white student population.  You can get the details here or read my earlier post here.

But then Mr. Kavet, the consultant , says 

In addition, there is no consistent measure of how well students are doing that can be compared from one school to another, Kavet said.

If that's true, then why are we spending money on the assessment tests we are giving to all the kids in all Vermont schools?  Now we find out that they don't tell us anything? 

Fortunately, the reporter, and consultant, remind us that

...the state also spends more per pupil than the vast majority of states and has the lowest pupil to teacher ratio in the nation.

If our average student performance is no better than the average non-minority student in the U.S. and we have no way to compare performance among schools, there's not much to brag about here.

And even if our performance is equal to the national average, consider this recent comment by former Federal Reserve Chairman Alan Greenspan in an excerpt from his new book, commenting on economic growth and his outlook for the nation's economic future:

Given that raw human intelligence is probably no greater today than in ancient Greece, our advancement will depend on additions to the vast heritage of human knowledge accumulated over the generations. A dysfunctional U.S. elementary and secondary education system [my emphasis] has failed to prepare our students sufficiently rapidly to prevent a shortage of skilled workers and a surfeit of lesser-skilled ones, expanding the pay gap between the two groups. Unless America's education system can raise skill levels as quickly as technology requires, skilled workers will continue to earn greater wage increases, leading to ever more disturbing extremes of income concentration. Education reform will take years...

...if it occurs at all.

Comments

[frank mazur]

Until parents take charge of education and force the legislature to level the playing field between the school board and teachers union education reform in Vermont will not happen.

[Anonymous]

Art: Curtis Hier says all we need to do is adjust classroom spending to 65% and everything will work out fine. Nice to see you talking about outcomes here.

[Curtis Hier]

Anonymous, let's get over our disagreement and move on. Obviously I have never said ALL we need to do is adjust classroom spending to 65 percent.

Voters have a certain amount of authority to affect change by voting budgets down. It is frustrating that it's difficult for voters to adopt benchmarks for spending and compare to different schools, because the numbers are kept so differently from district to district. I have recommended that percentages of current expeditures going to direct instruction and other categories be included in annual reports. The commissioner has it in his power to require it, but he's so far refused. If the information is good enough to see in the State's SASRS Report for 2006, then why can't we see it for our own schools for FY 2008?

Kavet and Rockler are correct in that supervisory union central offices take on certain expenses that in other districts are taken on by the individual schools. So I've recommended that we allocate SU expenses in calculating what each school spends for percentages in each category. Also, seeing what SU central offices spend has led me to support S.175 to require that spending to be voted on. Let's empower the voters by giving them good information. It's not ALL we need to do, but it's a start.

[Gregg Ballou]

Who cares how much education costs in VT. Think about the children, what about the children. And don't forget about the special needs children, but screw the gifted children.

Our lousy, but with a brand new bui8lding, elementary school is close to closing due to lack of enrollment and I hope it does before my kid starts school. Having a crappy local school certainly adds more impetus to moving and if I'm going to move I'm not stopping until I'm well over VT's border.

[Tim Diette]

Art, as usual I agree with your analysis. I would add one more piece to help the consumers of education evaluate the quality of schools: student growth or value-added measures of the test scores. The school level test scores are not an appropriate measure of the quality of the school unless all students enter school with the same test scores. Adding this bit of data would also help schools implement pay for performance. One final note, I don't have a problem with teaching to the test if it is a good test. I have not kept up with what tests are used in Vermont (as a resident of Virginia)...thanks...